Machine Learning (ML) is a complex task and it demands lots of resources to train and deploy ML model at scale for production. Training is only half the story; once you have trained your model, you typically want to use it to make predictions and there is a lot of focus on training. However when we talk to customers, running inference (prediction) in production represents the majority of the cost in ML workloads. In this session, we share how you can deploy an ML model using AWS Lambda and perform the inference via Amazon API Gateway in a cost effective and scalable manner. Learn how to build and deploy the whole application in an automated way, using AWS SAM (Serverless Application Model).
Suman Debnath, Principal Developer Advocate, AISPL